Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    parallel processing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    don
    0.70
     в
    0.68
     như
    0.68
     как
    0.64
    asionally
    0.62
    া
    0.62
    dong
    0.61
     नमू
    0.61
    står
    0.61
     некоторым
    0.61
    POSITIVE LOGITS
    지
    0.66
    ised
    0.65
    isations
    0.63
    수를
    0.58
    馬
    0.58
    車
    0.56
     conhecimento
    0.56
     exfoli
    0.56
    트
    0.55
    ிய
    0.54
    Act Density 0.006%

    No Known Activations